Enterprise business travel customized planning method and system based on big data

Abstract

The application discloses an enterprise business travel customized planning method and system based on big data, and relates to the technical field of big data.The application collects customer business travel data, extracts behavior intention, historical booking and browsing data, processes the behavior intention information, obtains key categories, carries out data feature extraction and clustering analysis, determines the best clustering number, analyzes group business travel preferences, establishes a dynamic law change graph based on a time stamp, uses a visualization tool to display a consumption mode, carries out comparative analysis according to a period, calculates business travel project correlation, sets a threshold to judge strong and weak correlation, combines real-time and historical data, adjusts the weight according to the value, regularly updates the historical data, and recommends personalized business travel products and optimizes the recommendation result based on user historical and real-time data through similarity calculation, so that a data-driven customized planning and personalized recommendation scheme is provided for enterprise business travel services, and multi-dimensional analysis and algorithm application are fused.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of big data, and particularly relates to an enterprise business travel customized planning method and system based on big data. BACKGROUND

[0002] In today's rapidly changing business environment, enterprise business travel planning and management are increasingly complex and critical. Traditional methods rely on manual experience and static data, making it difficult to capture subtle changes in market dynamics and customer needs.

[0003] With the development of big data technology, big data-based business travel customized planning methods have emerged to provide personalized, efficient and accurate solutions. The diversity and complexity of enterprise business travel activities require higher levels of planning. Different enterprises, departments and employees have significant differences in business travel needs, and traditional methods are difficult to cope with. Big data technology can analyze large amounts of customer data and reveal demand and preference patterns to provide a scientific basis for planning. Changes in market environment also require business travel planning to have higher flexibility and adaptability. Business travel costs are affected by a variety of factors and fluctuate, and customer needs change with market trends. Big data technology can monitor market dynamics in real time to help enterprises adjust strategies and respond to changes. In addition, personalized services have become an important trend in modern business travel services. Traditional standardized services are difficult to meet customer needs, and big data technology can mine customer historical behavior and preferences to provide customized solutions and improve customer experience and satisfaction.

[0004] In summary, big data-based enterprise business travel customized planning methods are introduced to address the diversity, complexity and market changes of business travel activities and meet the needs of personalized services. By deeply mining customer data, the application provides accurate, efficient and personalized business travel solutions to improve business travel management and market competitiveness. SUMMARY

[0005] To overcome the shortcomings and deficiencies of the prior art, the first object of the application is to provide a big data-based enterprise business travel customized planning method, and the second object of the application is to provide a big data-based enterprise business travel customized planning system.

[0006] The first object of the application adopts the following technical solutions:

[0007] The big data-based enterprise business travel customized planning method has the following process:

[0008] Step 1: Collect customer data related to business travel and preprocess the data to extract the behavior intention information, historical booking information and historical browsing big data of target users within a predetermined time period.

[0009] Step two, feature extraction and cluster analysis are performed on the preprocessed data, and frequent behavior sequence mining processing is performed on the processed behavior intention information, and the frequent behavior intention sequence and its support are output;

[0010] Step three, a dynamic regular change graph is established, and the consumption records and operation data of the customer are rearranged and analyzed based on the timestamp;

[0011] Step four, the correlation between business travel projects is calculated, and the business travel project combination with strong correlation is identified;

[0012] Step five, the real-time standard data set is merged with the customer historical data to form new historical data, and the weight proportion is adjusted according to the value of the data, and the business travel customization planning is carried out;

[0013] Step six, using collaborative filtering recommendation algorithm, combining time series analysis, personalized business travel scheme is recommended for customers.

[0014] Preferably, the data processing further comprises preprocessing the behavior intention information to eliminate noise and outliers to obtain behavior intention data, removing trip time conflicts and price abnormal data;

[0015] The behavior intention information is processed to obtain key categories, including business meeting category, training category and investigation category, and the segmentation process is as follows: the behavior intention data is sorted according to time sequence to obtain primary behavior intention sequence, each behavior intention node of the user has its corresponding position in the sequence, and the behavior intention sequence represents the behavior intention of the user within a certain timestamp; the behavior intention data is sorted according to time sequence to obtain primary behavior intention sequence Y, each behavior intention node of the user has its corresponding position in the sequence, and the behavior intention sequence represents the behavior intention of the user within a certain timestamp;

[0016] The primary behavior intention sequence set Y=(y1,y2,y3,y4,...,ym), where m is a positive integer, a corresponding point is created for each behavior intention ym in the sequence, and a corresponding behavior intention identifier, timestamp and page ID are added to retain the context information of each point, and the created points are sorted using timestamp to obtain the intermediate behavior intention sequence set X=(x1,x2,x3,x4,...,xm), where m is a positive integer;

[0017] Any point xm in the space is taken as the center of a circle, a radius r is set to form a circular region, and the set of all points in this circular region is marked as neighborhood B r (xm),B r(xm) = {xn∈D│dist(xm,xn)≤r}; where dist(xm,xn) represents the distance between xm and xn;

[0018] The minimum number of samples in the neighborhood is denoted as MinPts;

[0019] Randomly select a point xm from set X, and determine whether |B| r Is (xm)| greater than or equal to MinPts, when |B r If (xm)|≥MinPts, then xm is determined to be a seed point and added to the seed set Z;

[0020] Randomly select a seed point xn from the seed set Z, and add all points density-reachable from it to a new set C1, forming the first key category. Density reachability is defined as follows: if xn is in the neighborhood of xm, and xm is a seed point, then xn is density-reachable from xm. If there exist a1, a2, ..., a... n Where a1 = xm, a n = xn, and a i+1 By a i If the density is directly accessible, then xn can be reached from the density of xm;

[0021] Continue visiting the next point in set X, repeating the above steps until all points in the dataset have been processed and the key categories are obtained. Points not included in the key categories are marked as noise and deleted. The key categories obtained include business meeting category, training category, and inspection category.

[0022] Preferably, the data feature extraction and cluster analysis steps include: performing cluster analysis using the K-means algorithm, calculating the sum of squared clustering errors (SSE) for different K values ​​and plotting the curve of SSE versus K, and determining the optimal number of clusters K according to the elbow rule, wherein the formula for calculating SSE is... Where K is the number of clusters; C i Let x be the i-th cluster; x be the data point at the cluster center; μ i Let K be the centroid of the i-th cluster. Then, plot the curve of SSE as a function of K, and find the inflection point on the curve, i.e. the point where the sum of squared errors begins to decrease slowly, as the optimal K value. Describe the business travel preferences of each group based on the clustering results, analyze the attribute values ​​of the cluster center, and identify the destinations frequently visited by different groups and their preferred accommodation standards.

[0023] Preferably, the steps for establishing and analyzing the dynamic pattern change graph include: rearranging customer consumption records and operation data based on timestamps to establish a dynamic pattern change graph over a time span; using visualization tools to present the graph, showing changes in customer consumption patterns at different time periods, analyzing changes in customer business travel characteristics and preferences at different time periods, and marking peak and off-peak periods; setting a time period T, conducting comparative analysis by interval, comparing the differences in peak periods, off-peak periods, and business travel characteristics within different time intervals, and analyzing their impact on customer behavior and market trends.

[0024] Preferably, the steps for calculating the correlation of business travel projects include: for each customer's real-time standard dataset and historical data, calculating the correlation between preceding and subsequent business travel projects after sorting by time; applying the FP-Growth algorithm to mine frequent itemsets and generate association rules, analyzing the project correlation between two adjacent business travel records; setting a correlation threshold, and judging the correlation strength based on support, confidence, and lift, wherein the formula for calculating support is... Where σ(X∪Y) is the number of transactions containing itemsets X and Y; N is the total number of transactions, and the confidence score is calculated using the following formula: Where σ(X) is the number of transactions containing itemset X, and the lift is calculated using the formula: We explicitly define a support level greater than 0.5 and a confidence level greater than 0.7 as the criteria for judging a strong association.

[0025] Preferably, the step of combining real-time data with historical data includes: acquiring real-time standard datasets and customer historical data, merging the two to form new historical data; calculating a value score based on the data's time value, product association value, and continuity, using the formula Value = a1 × Time + a2 × Association + a3 × Continuity; where Time is the data's time value; Association is the product association value included in the data; and Continuity is the continuity, with the weights of the data's time value, product association value, and continuity in the value score being a1, a2, and a3, respectively, and valueless data having a weight of 0 in the original historical data; adjusting the weight ratios based on the value score, periodically updating the historical dataset, adding the latest transaction records, and recalculating the value score for customized business travel planning.

[0026] Preferably, the collaborative filtering recommendation algorithm steps include: based on the user's historical business travel records and real-time behavioral data, using collaborative filtering technology to predict business travel products that the user may be interested in; finding similar user groups by calculating the similarity between users; and recommending business travel products to the target user based on the historical behavior of similar users, wherein the formula for calculating cosine similarity is: Where A and B are the behavior vectors of two users; n is the vector dimension.

[0027] Preferably, the time series analysis step includes: using the ARIMA model to analyze users' historical consumption data; understanding the trend of user behavior over time; and further optimizing the recommendation results based on this trend.

[0028] The second objective of this invention is achieved through the following technical solution:

[0029] A big data-based enterprise business travel customization planning system is used to implement a big data-based enterprise business travel customization planning methodology. The system includes:

[0030] Data Acquisition and Processing Module: Responsible for collecting business travel-related data from multiple sources and preprocessing the data;

[0031] Customer segmentation module: Based on the preprocessed data, clustering algorithms are used to divide customers into different groups, and detailed feature descriptions are provided for each group;

[0032] Dynamic Pattern Analysis Module: Based on timestamps, the module rearranges customer consumption records and operation data to create a dynamic pattern change graph and performs data analysis.

[0033] Business travel project correlation calculation module: Calculates the correlation between customer business travel projects and identifies combinations of business travel projects with strong correlation;

[0034] Data merging and customized business travel planning module: Merges real-time standard datasets with customer historical data to form new historical data, and adjusts the weight ratio according to the value of the data to carry out customized business travel planning;

[0035] Personalized recommendation module: Utilizing collaborative filtering recommendation algorithms and time series analysis technology, it recommends personalized business travel solutions to customers.

[0036] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0037] 1. This invention identifies the business travel preferences of different groups (e.g., high-star hotels and frequent business trips, or budget accommodations and infrequent business trips) through user clustering analysis, and combines this with collaborative filtering recommendation algorithms (e.g., cosine similarity) to find similar user groups, providing users with tailored services. Furthermore, it uses the ARIMA model to predict future consumption trends, making recommendations more aligned with changing user needs.

[0038] 2. This invention utilizes dynamic pattern change graphs and interval comparison analysis to allow businesses to clearly observe changes in customer consumption patterns, peak and off-peak periods, thereby flexibly adjusting business travel strategies. By calculating the correlation of business travel projects and setting clear thresholds and indicators, resource allocation is optimized. The combination of real-time and historical data, along with a weighting adjustment mechanism, ensures that decisions are based on the latest and most valuable data, improving service competitiveness.

[0039] 3. This invention continuously updates the customer preference model by periodically updating historical datasets and recalculating value scores. The processing of behavioral intent information includes preprocessing, region segmentation, and frequent behavior sequence mining, helping businesses gain a deeper understanding of user behavior patterns. Based on these patterns and user feedback, services are continuously improved, such as providing timely recommendations of discounted tickets to customers who prefer specific regions, enabling dynamic adjustment and optimization of services. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 The flowchart of the enterprise business travel customization planning method based on big data of the present invention is shown;

[0042] Figure 2 This invention illustrates a module diagram of the enterprise business travel customization planning system based on big data.

[0043] Figure 3 A flowchart of step two of the present invention is shown. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more exemplary embodiments. Numerous specific details are provided in the following description to give a full understanding of exemplary embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, steps, etc., can be employed. In other instances, well-known structures, methods, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0046] Example 1:

[0047] See Figure 1 As shown in this embodiment, the process of the enterprise business travel customization planning method based on big data is as follows:

[0048] Step 1: Collect customer data related to business travel, preprocess the collected data, and extract behavioral intent information, historical booking information, and historical browsing big data of target users' business travel behavior data of candidate enterprises within a preset time period.

[0049] We collect business travel-related data from sources such as online booking platforms, customer feedback forms, and social media, including but not limited to spending amounts (such as airfare and hotel costs), spending frequency (number of trips), and product types (destination, accommodation standard, and mode of transportation).

[0050] For example, you can obtain airfare from airline booking systems, accommodation fees from hotel management systems, and feedback ratings from customer satisfaction surveys.

[0051] Use web analytics tools (such as Google Analytics) to track customer online behavior on online booking platforms, recording customer clickstream data. Operational data includes customer clicks and browsing history, monitored page visits, and browsing duration. For example, this includes the time users spend on the booking platform and the number of pages viewed.

[0052] Simultaneously, behavioral intent information is extracted, including itinerary information (such as travel date and trip duration), service selection information (such as transportation mode selection and hotel class selection), and travel scenario information (such as business travel and leisure travel); historical booking information includes booking service information (such as booked hotels, flights, car rentals, etc.), booking frequency information, and booking service details (such as room type and flight class). Timestamps are the time information corresponding to the customer's business travel activities (such as booking time and travel time), ensuring that all collected data is accompanied by accurate timestamps.

[0053] For example, each booking operation generates a timestamp that marks the exact time the operation occurred.

[0054] Ensure all data processing activities comply with relevant privacy regulations (such as GDPR) and take necessary encryption measures to protect personal information. Clearly inform users of the purpose of data use and obtain their consent.

[0055] Step 2: Data feature extraction and cluster analysis. At the same time, after processing the behavioral intent information, frequent behavior sequence mining is performed to output frequent behavior intent sequences and their support.

[0056] See Figure 3 As shown, the specific process is as follows:

[0057] S21. Use the K-means algorithm for cluster analysis and determine the optimal number of clusters K using the elbow rule.

[0058] First, calculate the sum of squared clustering errors (SSE) for different K values, then plot the curve of SSE changing with the K value. Find the inflection point on the curve, i.e., the point where the sum of squared errors begins to decrease slowly, as the optimal K value.

[0059] The formula for calculating SSE is as follows: Where K is the number of clusters; C i Let x be the i-th cluster; x be the data point at the cluster center; μ i Let K be the centroid of the i-th cluster; then, plot the curve of SSE as a function of K, and find the inflection point on the curve, i.e. the point where the sum of squared errors begins to decrease slowly, as the optimal K value.

[0060] Based on the clustering results, each group is described in detail to identify the business travel preferences of different groups (such as frequently visited destinations, preferred accommodation standards, etc.). The attribute values ​​of each cluster centroid are analyzed. For example, some groups may prefer high-star hotels, while others may prefer budget accommodations.

[0061] S22. Process the behavioral intent information.

[0062] S201. Preprocess the behavioral intent information by cleaning and preprocessing the data to remove noise and outliers. The above processing yields behavioral intent data. Clean the various behavioral intent information of users in business travel to remove unreasonable data, such as data on conflicting travel times and abnormal service selections (such as data on abnormally low or high prices), thus obtaining behavioral intent data.

[0063] 202. Process the behavioral intent information to obtain key categories, including business meeting category, training category, and inspection category;

[0064] Specifically, the behavioral intent data is sorted according to time order to obtain a primary behavioral intent sequence. Each user's behavioral intent node has its corresponding position in the sequence, and the behavioral intent sequence represents the user's behavioral intent within a certain timestamp. The behavioral intent data is sorted according to time order to obtain a primary behavioral intent sequence Y. Each user's behavioral intent node has its corresponding position in the sequence, and the behavioral intent sequence represents the user's behavioral intent within a certain timestamp.

[0065] The set of primary behavioral intent sequences Y = (y1, y2, y3, y4, ..., ym), where m is a positive integer, creates a corresponding point for each behavioral intent ym in the sequence, and adds the corresponding behavioral intent identifier, timestamp, and page ID to preserve the context information of each point. The created points are sorted using the timestamps to obtain the set of intermediate behavioral intent sequences X = (x1, x2, x3, x4, ..., xm), where m is a positive integer;

[0066] Let any point xm in space be the center of a circle, and set the radius to r to form a circular region. The set of all points within this circular region is labeled as the neighborhood B. r (xm),

[0067] B r (xm)={xn∈D│dist(xm,xn)≤r};

[0068] Where dist(xm,xn) represents the distance between xm and xn;

[0069] The minimum number of samples in the neighborhood is denoted as MinPts;

[0070] Randomly select a point xm from set X, and determine whether |B| r Is (xm)| greater than or equal to MinPts, when |B r If (xm)|≥MinPts, then xm is determined to be a seed point and added to the seed set Z;

[0071] Randomly select a seed point xn from the seed set Z, and add all points density-reachable from it to a new set C1, forming the first key category. Density reachability is defined as follows: if xn is in the neighborhood of xm, and xm is a seed point, then xn is density-reachable from xm. If there exist a1, a2, ..., a... n Where a1 = xm, a n = xn, and a i+1 By a i If the density is directly accessible, then xn can be reached from the density of xm;

[0072] Continue visiting the next point in set X and repeat the above steps until all points in the dataset have been processed and the key categories are obtained. Points not included in the key categories are marked as noise and deleted.

[0073] Key categories include business meetings, training, and study tours.

[0074] S203. Use the Apriori algorithm to mine frequent behavioral intent sequences for key categories, output frequent behavioral intent sequences and their support, set a support threshold, remove behavioral intent sequences below the support threshold, retain behavioral intent sequences above the support threshold and their support, and mark them as frequent behavioral intent sequences. Here, support represents the frequency of a user's behavioral intent sequence in the dataset.

[0075] Step 3: Establishment and analysis of dynamic pattern change diagram.

[0076] First, based on timestamps, customer consumption records and operation data are rearranged and a dynamic pattern change graph is established. For each business travel booking consumption record and operation data on the booking platform, they are arranged in chronological order according to timestamps, presenting the continuity and variability of customer business travel behavior in the time dimension.

[0077] Next, visualization tools are used to present the established dynamic patterns of change. Charts are used to show changes in customer spending patterns over different time periods, creating line graphs of total business travel spending for each month or quarter. Based on the fluctuations in spending amounts in the graphs, peak and off-peak periods of consumption are identified. Simultaneously, through in-depth analysis of data from different time periods, the characteristics and preferences of customers in business travel are summarized, including a preference for high-end hotels during certain periods, while prioritizing convenient and cost-effective transportation during other periods.

[0078] Finally, a time period T is set and a comparative analysis is conducted across different time intervals. By dividing the time period into different intervals by quarter or month, data within each interval is compared. The comparison includes differences in the timing, duration, and intensity of peak and off-peak periods. It also focuses on changes in business travel characteristics, such as destination preferences and mode of transportation preferences, across different intervals. The impact of these differences on customer behavior is analyzed, including how price fluctuations during peak periods lead customers to book earlier or later, or change the type of products booked. The impact on market trends is also studied, including how changes in the popularity of certain destinations during specific time periods affect the supply and pricing strategies of related business travel products. This comprehensive analysis provides important guidance for companies to adjust their customized business travel planning strategies, enabling them to better adapt to market changes and meet customer needs. Step Four: Calculation of Business Travel Project Correlation.

[0079] S41, Correlation Calculation.

[0080] For each customer's real-time standard dataset and historical data, after sorting by time, the correlation between preceding and subsequent business travel projects is calculated. The correlation between projects in two adjacent business travel records is analyzed.

[0081] Implementation details: The FP-Growth algorithm is applied to mine frequent itemsets and generate association rules. For example, if it is found that customers often book the same hotel chain after booking a flight, then a strong association can be considered between the two items.

[0082] S42. Association judgment.

[0083] A correlation threshold is set. If the calculated correlation exceeds this threshold, the two business travel projects are considered to have a strong correlation; otherwise, they are considered to have a weak correlation. The correlation indicators include support, confidence, and lift. The formula for calculating support is... Where σ(X∪Y) is the number of transactions containing itemsets X and Y; N is the total number of transactions; the confidence score is calculated using the following formula: Where σ(X) is the number of transactions containing itemset X; the formula for calculating lift is...

[0084] Define clear threshold criteria, such as support greater than 0.5 or confidence greater than 0.7, to filter out association rules that have practical significance.

[0085] Example: Analysis revealed that when customers book flights to Shanghai, there is an 80% probability that they will choose to book hotels of a specific brand. Therefore, corresponding customized business travel planning strategies can be set up.

[0086] Step 5: Combine real-time data with historical data.

[0087] S51, Data Merging and Weight Adjustment.

[0088] Acquire real-time standard datasets and customer historical data, merge the real-time standard datasets with the original historical data to form new historical data, and adjust the weight ratios according to the value scores of the real-time standard datasets and customer historical data.

[0089] The value score is a comprehensive evaluation based on the data's time value, the associated product values, and continuity. The formula is: Value = 0.3 × Time + 0.2 × Association + 0.5 × Continuity; where Time represents the data's time value, Association represents the associated product values, and Continuity represents continuity. The weighting is adjusted according to the value score: the data's time value accounts for 0.3%, the associated product values ​​account for 0.2%, continuity accounts for 0.5%, and data with no value has a weight of 0 in the original historical data. Regarding the data's time value: newer data is generally considered more valuable because it better reflects current user preferences; this indicator is quantified by calculating the time difference between the data's generation date and the present.

[0090] Product association values ​​included in the data: These measure the strength of the association between different products or services involved in the data; for example, if a user frequently stays at the same hotel chain after booking a flight, it indicates a strong association between the flight and the hotel.

[0091] Continuity: This refers to the consistency and repeatability of user behavior. Highly consistent behavior means that users tend to repeat certain specific behavioral patterns, which is very important for predicting future behavior.

[0092] For example: Developing a business travel plan for a client who is based in Beijing but occasionally needs to travel to Shanghai for meetings:

[0093] If it is found that the customer has chosen a particular airline and a specific brand of hotel on their most recent trips to Shanghai, and this choice has continued for a period of time (high continuity), then this data will have a high value score.

[0094] If further analysis shows that after booking a ticket with this airline, the customer almost always books the same hotel chain (high product association value), then this data will also receive a higher value score.

[0095] Bookings made in recent months are more important than those made a few years ago (time value).

[0096] S52, Customized Business Travel Planning.

[0097] Based on new historical data, dynamic pattern change graphs are generated and the correlation of business travel projects is calculated. The historical dataset is updated regularly, the latest transaction records are added, and the value score is recalculated to carry out customized business travel planning.

[0098] Implementation details: Historical datasets are updated regularly, incorporating the latest transaction records and recalculating the value score for customized business travel planning. The value score is defined by comprehensively evaluating the data's time value, the product relevance it contains, and its continuity. The value score considers factors such as data freshness, product relevance, and continuity to comprehensively assess the data's importance.

[0099] Example: With the continuous addition of new data, customer preference models can be updated in a timely manner, and business travel customization planning strategies can be adjusted according to the latest trends. For example, if it is found that flight prices in a certain region have recently decreased, discounted tickets can be recommended to customers who prefer that region.

[0100] The time value of the data accounts for 0.3% of the value score.

[0101] The proportion of product-related value to the total value is 0.2%.

[0102] Continuity accounts for 0.5% of the value score.

[0103] The weight of the valueless data in the original historical data is 0.

[0104] Step 6: Use collaborative filtering recommendation algorithms to recommend personalized business travel plans to customers.

[0105] S61, Collaborative Filtering Recommendation.

[0106] Based on users' historical business travel records and real-time behavioral data, collaborative filtering techniques are used to predict business travel products that users may be interested in. For example, by calculating the similarity between users (such as cosine similarity), user groups similar to the target user are identified, and business travel products are recommended to the target user based on the historical behavior of similar users.

[0107] The formula for calculating cosine similarity is:

[0108] Where A and B are the behavior vectors of two users; n is the vector dimension.

[0109] Example: For a client who is based in Beijing but occasionally needs to travel to Shanghai for meetings, the system can recommend cost-effective round-trip airfares and suitable hotel packages based on their historical records and current market conditions.

[0110] S62. Combine with time series analysis.

[0111] By combining time series analysis, we can understand the changing trends of user behavior over time and further optimize recommendation results. For example, using the ARIMA model to perform time series analysis on users' historical consumption data can predict future consumption trends, thereby more accurately recommending business travel products to users.

[0112] The beneficial effects of this embodiment are as follows: Big data analysis enables a deep understanding of customers' business travel behavior, which not only improves the quality of personalized services but also accurately predicts future customer needs. Cluster analysis is used to identify the preferences of different user groups, dynamic pattern change graphs are used to capture changes in consumption patterns, and association rule mining enhances the relevance of product recommendations.

[0113] Example 2:

[0114] See Figure 2 As shown, the enterprise business travel customization planning system based on big data in this embodiment includes a data acquisition and processing module, a customer group segmentation module, a dynamic pattern analysis module, a business travel project correlation calculation module, a data merging and business travel customization planning module, and a personalized recommendation module.

[0115] Data Acquisition and Processing Module: Responsible for collecting business travel-related data from multiple sources and preprocessing the data to ensure its quality and availability.

[0116] Customer segmentation module: Based on the preprocessed data, clustering algorithms are used to divide customers into different groups, and detailed feature descriptions are provided for each group.

[0117] Dynamic Pattern Analysis Module: Based on timestamps, the module rearranges customer consumption records and operation data to create a dynamic pattern change graph and performs data analysis.

[0118] Business travel project correlation calculation module: Calculates the correlation between customer business travel projects and identifies combinations of business travel projects with strong correlation.

[0119] Data merging and customized business travel planning module: Merges real-time standard datasets with customer historical data to form new historical data, and adjusts the weight ratio according to the value of the data to carry out customized business travel planning.

[0120] Personalized recommendation module: Utilizing collaborative filtering recommendation algorithms and time series analysis technology, it recommends personalized business travel solutions to customers.

[0121] The beneficial effects of this embodiment are: it automates the entire process from data collection to personalized recommendations, effectively improving the efficiency and accuracy of business travel planning. The system can deeply understand customer needs, provide customized services, enhance customer experience, and bring higher business travel management benefits to enterprises.

[0122] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0123] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A big data-based method for customized business travel planning, characterized in that: The method flow is as follows: Step 1: Collect customer data related to business travel, and preprocess the data to extract the target user's behavioral intent information, historical booking information, and historical browsing big data within a preset time period; Step 2: Perform feature extraction and cluster analysis on the preprocessed data, and simultaneously perform frequent behavior sequence mining on the processed behavioral intent information to output frequent behavior intent sequences and their support. Step 3: Establish a dynamic pattern change chart, and rearrange and analyze customer consumption records and operation data based on timestamps; The steps for establishing and analyzing the dynamic pattern change chart include: First, rearranging customer consumption records and operation data based on timestamps to establish a dynamic pattern change chart. For each business travel booking's consumption records and operation data on the booking platform, they are arranged in chronological order according to timestamps, presenting the continuity and variability of customer business travel behavior over time. Next, using visualization tools, the established dynamic pattern change chart is presented, displaying changes in customer consumption patterns at different time periods through charts, drawing line graphs of total business travel consumption for each month or quarter, and determining peak and trough periods based on fluctuations in consumption amounts in the charts. Simultaneously, by analyzing data from different time periods... In-depth analysis revealed the characteristics and changing preferences of business travelers, including a preference for high-end hotels at certain times and a greater emphasis on convenient and cost-effective transportation at other times. Finally, a time period T was established and inter-period comparative analysis was conducted. By dividing the time period into quarterly or monthly intervals, data within each interval was compared. The comparisons included differences in the timing, duration, and intensity of peak and off-peak periods. It also examined changes in business travel characteristics, such as destination preferences and mode of transportation preferences, across different intervals, analyzing the impact of these differences on customer behavior, including how price fluctuations during peak periods led customers to book earlier or later, or change the type of products booked. Step 4: Calculate the correlation between business travel items and identify combinations of business travel items with strong correlations; the steps for calculating the correlation of business travel items include: for each customer's real-time standard dataset and historical data, sort by time and calculate the correlation between business travel items before and after; apply the FP-Growth algorithm to mine frequent itemsets and generate association rules, and analyze the item correlation between two adjacent business travel records; Step 5: Merge the real-time standard dataset with customer historical data to form new historical data, and adjust the weight ratios according to the data's value score for customized business travel planning. The steps for combining real-time and historical data include: acquiring the real-time standard dataset and customer historical data, merging them to form new historical data; calculating the value score based on the data's time value, product association value, and continuity, using the following formula: ;in, The time value of the data; The data contains product-related values. For continuity, the weights of the data's time value, product association value, and continuity in the value score are as follows: , , Valueless data has a weight of 0 in the original historical data; the weight ratio is adjusted according to the value score, the historical dataset is updated regularly, the latest transaction records are added, and the value score is recalculated for customized business travel planning; Step Six: Utilize collaborative filtering recommendation algorithms, combined with time series analysis, to recommend personalized business travel plans to customers.

2. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, The data processing in step one also includes preprocessing behavioral intent information, eliminating noise and outliers to obtain behavioral intent data, and removing data on trip time conflicts and price anomalies. The behavioral intent information is processed to obtain key categories, which include business meeting category, training category and inspection category. The segmentation process is as follows: the behavioral intent data is sorted according to time order to obtain a primary behavioral intent sequence. Each user's behavioral intent node has its corresponding position in the sequence. The behavioral intent sequence represents the user's behavioral intent within a certain time stamp. The behavioral intent data is sorted chronologically to obtain a primary behavioral intent sequence. Each user's behavioral intent node has its corresponding position in the sequence, and the behavioral intent sequence represents the user's behavioral intent within a certain timestamp; Primary behavioral intention sequence set ,in, A positive integer representing the intent of each action in the sequence. Create a corresponding point and add a corresponding behavioral intent identifier, timestamp, and page ID to preserve the context information of each point. Sort the created points using timestamps to obtain a set of intermediate behavioral intent sequences. ,in, Let be a positive integer; let any point in space be... Using a circle as its center and a radius of r, a circular region is formed. The set of all points within this circular region is labeled as its neighborhood. , ;in, express and The distance between them; the minimum number of samples in the neighborhood is marked as MinPts; from the set Randomly select a point ,judge Is it greater than or equal to MinPts, when Then determine Add the seed point to the seed set. middle; From seed set Randomly select a seed point Add all points that are density-reachable to this set to a new set C1, forming the first key category; where density-reachability is defined as follows: if exist Within its neighborhood, and If it is a seed point, then Depend on Density reaches directly, if it exists , ,..., ,in = , = ,and Depend on Density reaches directly, then Depend on Density can be achieved; Continue accessing the collection For the next point in the dataset, repeat the above steps until all points in the dataset have been processed to obtain the key categories. Points not included in the key categories are marked as noise and deleted. The key categories obtained include business meeting category, training category, and inspection category.

3. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, The data feature extraction and cluster analysis steps include: performing cluster analysis using the K-means algorithm, calculating the sum of squared clustering errors (SSE) for different K values ​​and plotting the curve of SSE versus K, and determining the optimal number of clusters K based on the elbow rule, where the formula for calculating SSE is... ;in, The number of clusters; For the first One cluster; Data points that serve as cluster centers; For the first The centroids of each cluster are determined; then, the curve of SSE versus K value is plotted, and the inflection point on the curve, i.e. the point where the sum of squared errors begins to decrease slowly, is taken as the optimal K value; the business travel preferences of each group are described based on the clustering results, and the attribute values ​​of the cluster center points are analyzed to identify the destinations frequently visited by different groups and their preferred accommodation standards.

4. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, The steps for calculating the correlation of business travel projects also include setting a correlation threshold and judging the correlation strength based on support, confidence, and lift; specifying that support greater than 0.5 and confidence greater than 0.7 are the criteria for judging strong correlation.

5. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, In the step of combining real-time data with historical data, the time value of the data is quantified by calculating the time difference between the data generation date and the present. Updated data is considered to be more valuable and better reflects the user's current preferences. ;in, It is the time difference between the data generation time and the current time; It is the decay rate parameter; Product association values ​​included in the data: These measure the strength of the association between different products or services involved in the data; for example, if a user frequently stays at the same hotel chain after booking a flight, it indicates a strong association between the flight and the hotel. Continuity: refers to the repetitiveness of user behavior. High continuity of behavior means that users tend to repeat certain specific behavioral patterns. ;in, Representing the The frequency of repetition of a specific behavioral pattern, that is, the number of times the behavior occurs; This indicates the relative importance weight of the behavior pattern among all behaviors. This represents the total number of user actions during the observation period; It is a time decay factor, used to measure whether the importance of a particular behavior decreases over time. ,in It is the time difference between the current time and the occurrence of the action. It is the decay rate; relative importance weight. The process of obtaining these metrics is as follows: Each behavioral value indicator is determined, including consumption amount, contribution to corporate profits, and enhancement of brand image; different behavioral value indicators are then normalized; for each indicator… The normalized value ;in It is a behavioral pattern Indicators The original value, and These are indicators Find the minimum and maximum values ​​across all behavioral patterns; calculate the overall value score for each behavioral pattern. ;in, It is an indicator Finally, the overall value scores of all behavioral patterns are normalized to obtain relative importance weights. ;in, It represents the total number of behavioral patterns.

6. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, The collaborative filtering recommendation algorithm includes the following steps: based on users' historical business travel records and real-time behavioral data, using collaborative filtering technology to predict business travel products that users may be interested in; finding similar user groups by calculating the similarity between users; and recommending business travel products to target users based on the historical behavior of similar users, wherein the formula for calculating cosine similarity is... ;in, and For the behavior vectors of the two users; For vector dimensions.

7. The enterprise business travel customization planning method based on big data according to claim 1, characterized in that, The time series analysis step includes: using the ARIMA model to analyze users' historical consumption data; understanding the trend of user behavior over time; and further optimizing the recommendation results based on this trend.

8. A big data-based enterprise business travel customization planning system, used to implement the big data-based enterprise business travel customization planning method as described in claim 1, characterized in that, The system includes: Data Acquisition and Processing Module: Responsible for collecting business travel-related data from multiple sources and preprocessing the data; Customer Group Segmentation Module: Based on the preprocessed data, uses clustering algorithms to segment customers into different groups and provides detailed characteristic descriptions for each group; Dynamic Pattern Analysis Module: Based on timestamps, rearranges customer consumption records and operation data to establish dynamic pattern change graphs and performs data analysis; Business Travel Project Correlation Calculation Module: Calculates the correlation between customer business travel projects and identifies combinations of business travel projects with strong correlations; Data Merging and Customized Business Travel Planning Module: Merges real-time standard datasets with customer historical data to form new historical data, adjusts weight ratios according to data value, and performs customized business travel planning; Personalized Recommendation Module: Uses collaborative filtering recommendation algorithms and time series analysis technology to recommend personalized business travel solutions to customers.

Images